GLM-TTS Use Cases & Resources
Common production scenarios and official download links.
Education
Mixed Chinese/English, formulas, and polyphones with phoneme control.
Audiobooks & Storytelling
Multi-role narration with a wide emotional range (crying, laughing, shouting).
Smart Customer Service
Warm, professional speech with stable prosody even with variable inserts.
Zero-shot Voice Cloning
Clone timbre and prosody from ~3 seconds of prompt audio.
Emotion Control (GRPO RL)
Nuanced emotions (happy/sad/angry) plus natural laughter/breathing.
Phoneme-in Pronunciation
Hybrid phoneme + text input for polyphones and rare words.
Model Weights (Hugging Face)
Download checkpoints from zai-org/GLM-TTS.
Model Weights (ModelScope)
Recommended mirror for users in China.
Gradio Web UI
Run an interactive web demo locally via tools/gradio_app.py.